WARNING: Version 5.0 of the Elastic Stack has passed its EOL date.
This documentation is no longer being maintained and may be removed. If you are running this version, we strongly advise you to upgrade. For the latest information, see the current release documentation.
Marvel Release Notes (Pre-5.0)
editMarvel Release Notes (Pre-5.0)
editThis section summarizes the Monitoring changes in each release. For information about upgrading X-Pack, see Upgrading X-Pack.
Change List
edit2.4.2
editNovember 22, 2016
Enhancements
- Compatibility with Elasticsearch 2.4.2.
2.4.0
editAugust 31, 2016
Enhancements
- Node transport addresses are now consistently reported using only their IP address and port, rather than a subset reporting their resolved hostname, IP address, and port.
Bug Fixes
- Using the time picker to select clusters now honors the full time range, rather than only clusters that sent updates within the last 10 minutes.
2.3.5
editAugust 3, 2016
Bug Fixes
- Fixed sorting by number of unassigned shards in the indices table.
- Fixed the shard allocation tables to render when shards are relocating.
- Fixed the tooltips that are supposed to show for relocating shards in the allocation table, and also added tooltips for shards of all states.
- Removed Google Analytics beacon, and fix the optional phone-home call to not block page rendering.
2.3.4
editJuly 7, 2016
Bug Fixes
- Using the time picker to select clusters now honors the full time range, rather than only clusters that sent updates within the last 10 minutes.
-
Phone home data omits
source_node
information that was being erroneously picked up from thecluster_info
field. - Shards in the Relocating state are no longer counted as Unassigned Shards.
- Removed Google Analytics Tag from UI.
- The UI rendering no longer blocks waiting for the phone home call to return.
- Fixed the "no data" page to run an automatic refresh cycle to look for data.
2.3.3
editMay 18, 2016
Bug Fixes
- Prevents exceptions from being logged if the node stats are temporarily empty.
2.3.2
editApril 26, 2016
Bug Fixes
-
Fixed a bug where non-elected master nodes all appeared as Offline in the node list.
You can workaround this in 2.3.0 and 2.3.1 by setting
marvel.node_resolver: name
inkibana.yml
. - Introduced the timepicker controls in the clusters page so you can view snapshots Marvel data.
- Added a link to go back to the previous page from the License page.
2.3.1
editApril 4, 2016
Enhancements
- Adds support for Elasticsearch 2.3.1
2.3.0
editMarch 30, 2016
Bug Fixes
- Minor UI improvements.
Enhancements
- Adds support for Elasticssearch 2.3.0
2.2.1
editMarch 15, 2016
Bug Fixes
- Fix a bug when Monitoring is configured to export data over HTTP on a monitoring cluster protected by Security (with SSL).
2.2.0
editFebruary 2, 2016
Enhancements
- Enable access to the License page from the cluster status bar.
- Allow configuration of a monthly interval for Monitoring indices instead of limiting it to daily.
- Provide a sha1 hash for the Monitoring app download.
Bug Fixes
- Minor UI improvements.
2.1.2
editFebruary 2, 2016
Enhancements
- Adds support for Elasticsearch 2.1.2
2.1.1
editDecember 17, 2015
Bug Fixes
- Check the Kibana version to ensure compatibility.
- Respect opt-out for statistics reporting in all cases.
- Fixed transport address not showing on the node detail page.
- Fixed the table filtering on the node listing page.
2.1.0
editNovember 24, 2015
Enhancements
- Display a node type indicator on the Node Listing and Node Detail pages.
- Show the online/offline status of nodes.
- Remove the Time Picker from the Cluster Listing page.
Bug Fixes
- Multiple fixes for showing historical data.
- Filter out derivative calculations where the start and end times are incomplete time buckets.
- Cluster table sorting.
-
Use the new Elasticsearch
_field_stats
API to determine which Monitoring indices are needed to generate the stats.
2.0.1
editNovember 24, 2015
Enhancements
- Adds support for Elasticsearch 2.0.1.
2.0.0
editOctober 28, 2015
Breaking ChangesMonitoring 2.0.0 is a complete rewrite from Monitoring 1.3. If you are upgrading from Monitoring 1.3, note that
the Monitoring agent plugin only needs to be installed on the Elasticsearch nodes you are monitoring.
If you need to stop collecting data from a node, remove the plugin or set the xpack.monitoring.enabled
configuration parameter to false
. (The xpack.monitoring.agent.enabled
parameter is no longer supported.)
New Features
- Fully-integrated with Kibana 4.2 as a Kibana plugin.
- Supports Multiple clusters for Paid Subscriptions
- Reduced data footprint for Monitoring indices
- Simplified dashboards for Indices and Nodes
- Better support for large clusters